Multivariate prognostic evaluation of the mitotic activity index and fibrotic focus in node-negative invasive breast cancers

Baak, Jan P.A.; Colpaert, Cecile G.A.; van Diest, Paul J.; Janssen, Emiel; Diermen, Bianca van; Albernaz, Eliza; Vermeulen, Peter B.; Van Marck, Eric A.

doi:10.1016/j.ejca.2005.03.038

« Previous Next »European Journal of Cancer
Volume 41, Issue 14 , Pages 2093-2101, September 2005

Multivariate prognostic evaluation of the mitotic activity index and fibrotic focus in node-negative invasive breast cancers

Jan P.A. Baak
- Department of Pathology, Stavanger University Hospital, Armauer Hansensveg 20, P.O. Box 8100, 4068 Stavanger, Norway
- The Gade Institute, University of Bergen, Norway
- Free University Medical Centre, Amsterdam, The Netherlands
- Corresponding author. Tel.: +47 51 519534; fax: +47 51 519910.
Cecile G.A. Colpaert
- University Hospital, Antwerp, Belgium
Paul J. van Diest
- University Medical Center, Utrecht, The Netherlands
Emiel Janssen
- Department of Pathology, Stavanger University Hospital, Armauer Hansensveg 20, P.O. Box 8100, 4068 Stavanger, Norway
Bianca van Diermen
- Department of Pathology, Stavanger University Hospital, Armauer Hansensveg 20, P.O. Box 8100, 4068 Stavanger, Norway
Eliza Albernaz
- Department of Pathology, Stavanger University Hospital, Armauer Hansensveg 20, P.O. Box 8100, 4068 Stavanger, Norway
Peter B. Vermeulen
- University Hospital, Antwerp, Belgium
Eric A. Van Marck
- University Hospital, Antwerp, Belgium

Received 28 January 2005; received in revised form 9 March 2005; accepted 30 March 2005.

Abstract

We validated with univariate and multivariate (Cox) analysis, the prognostic value of the mitotic activity index (MAI), the fibrotic focus (FF) and other prognosticators in 448 patients with lymph node-negative (LN−) invasive breast cancer <55 years without adjuvant systemic treatment (72.5 months median follow-up, range 4–119). Of these patients, 24.8% developed distant and 1.6% loco-regional recurrence. FF showed excellent inter-observer reproducibility (κ=0.93). Strong prognosticators were MAI, grade, nuclear atypia, FF and the St. Gallen criterion (SG). The subgroup with excellent survival selected by SG was only 16% of all patients, implying over-treatment of more than 70% of all LN− patients when using SG as adjuvant therapy selection criterion. If MAI <10, 13% showed distant metastases, contrasting with 41% if MAI ⩾10. FF was prognostic in the ductal and mixed ductal cancers, but not in the lobular and other subtype cancers. Patients with invasive (mixed) ductal cancers with FF absent, FF<1/3 or FF>1/3 of the tumour area, had distant metastasis rates of 17%, 35% and 48%; in MAI<10 and FF absent, FF<1/3 or FF>1/3, metastasis rates were 11%, 13% and 42% and if MAI⩾10, metastasis rates were 31%, 48% and 50%, respectively. In the 12 patients with MAI<10 and a large FF>1/3, event-free survival was similar to patients with MAI⩾10. With multiple regression MAI<10 versus ⩾10 is the strongest prognosticator (also stronger than the SG). The FF may be important as it has additional prognostic value to the MAI in the small subgroup of invasive ductal or mixed-ductal breast cancer patients with combined MAI<10 and an FF>1/3 of the tumour area.